An air-ground cooperative unmanned aerial vehicle operation risk assessment method

Abstract

The application discloses a kind of air-ground coordinated unmanned aerial vehicle operation risk assessment method, facing unmanned aerial vehicle city operation scene, belong to unmanned aerial vehicle risk assessment field.First, based on system failure and track error probability density distribution, respectively calculate unmanned aerial vehicle operation system failure and collision failure probability;Second, based on kinematic model, solve the probability density distribution function of the predicted crash point of unmanned aerial vehicle under ballistic descent condition;Finally, use personnel damage, economic loss and noise influence multiple indexes to evaluate the operation risk of unmanned aerial vehicle.The application faces the city operation risk scene of fine modeling, can calculate the instantaneous operation risk of unmanned aerial vehicle, to provide corresponding support to its operation planning in city operation scene.

Description

Technical Field

[0001] This invention relates to the field of unmanned aerial vehicle (UAV) operation risk assessment technology, and specifically to an air-ground collaborative UAV operation risk assessment method. Background Technology

[0002] The operational risk of unmanned aerial vehicles (UAVs) refers to the potential harm that a UAV may cause to objects in its operating environment during operation. From an assessment perspective, it can be divided into two aspects: risk probability and severity. The occurrence of UAV operational risk follows an accident chain structure, where an in-flight failure occurs, leading to a crash and damage to the ground. Therefore, UAV operational risk assessment methods should combine the probability of in-flight failure, crash trajectory prediction, and impact damage severity to ultimately calculate the UAV operational risk that closely reflects actual operating conditions.

[0003] In recent years, research on the risk assessment of drone operations has focused on the study of ground impact risk, defining the probability of drone failure in the air as a fixed parameter and simplifying the fall trajectory to a vertical fall. The corresponding settings do not conform to the actual operating conditions.

[0004] In the low-altitude operating environment of drones, the dense obstacles, diverse population density, and complex ground conditions place higher demands on drone operational risk assessment methods. From the perspective of the probability of risk occurrence, in addition to system malfunctions, collisions with obstacles must also be considered as events that could cause drone in-flight failure. Secondly, the differences in population density necessitate the prediction of crash trajectories. Furthermore, from the perspective of the severity of the risk, in addition to personal injury, economic losses and noise impacts must also be considered. Summary of the Invention

[0005] To address the problems of the prior art, this invention proposes an air-ground collaborative UAV operation risk assessment method. Based on the limitations of complex low-altitude environments and UAV operation safety requirements, it can provide support for refined UAV risk assessment and provide a basis for verifying the rationality of UAV operation planning.

[0006] The technical solution for realizing the present invention is as follows:

[0007] A method for risk assessment of air-ground coordinated unmanned aerial vehicle (UAV) operations includes the following steps:

[0008] Step S1: Calculate the failure probability of the UAV operating system and the probability of collision failure based on the probability density function distribution of system failure and collision.

[0009] Step S2: Solve the probability density distribution function of the expected crash point of the UAV under ballistic descent using a kinematic model;

[0010] Step S3: Using an accident chain model, assess the instantaneous operational risk of the drone using multiple indicators.

[0011] Furthermore, step S1 specifically includes:

[0012] Step S11: Based on system reliability theory, set the failure probability of the UAV system. It follows a Weibull distribution, and its failure probability function is:

[0013]

[0014] in, This refers to the cumulative operating time of the drone; For proportional parameters, These are shape parameters, both determined by different stages of system operation;

[0015] Step S12: Obtain the initial nominal position of the UAV The trajectory error of the UAV is set to follow a Gaussian distribution, which is: ,in Let be the trajectory error values ​​along each direction of the aircraft's coordinate axes; then the position distribution of the UAV is:

[0016]

[0017] in, The nominal average value. This is the transformation matrix between the machine's coordinate axes and the global coordinate axes. The variance of the transformed probability distribution;

[0018] Step S13: Based on the above probability distribution definition, calculate the relative position of the UAV and the target obstacle:

[0019] in, The location of the drone. Location of the target obstacle. and These are the position transformation matrices for both. and These are the positional errors of the two, and These are the nominal positions of the two, respectively.

[0020] The probability distribution of their relative positions is obtained:

[0021]

[0022] in, The mean of relative positions. Let covariance be the probability distribution.

[0023] Step S14: Calculate the collision probability between the UAV and the target obstacle based on the three-dimensional Gaussian distribution integral.

[0024]

[0025] in, This refers to the cumulative operating time of the drone. For the integration region, Relative position The initial relative position mean, Let be the relative position covariance.

[0026] Furthermore, step S2 specifically includes:

[0027] Step S21: Obtain the initial nominal position of the UAV With running speed drone quality Calculate the drag coefficient ;

[0028]

[0029] in, air density, For windward area, The air drag coefficient is given; and the maximum descent speed of the UAV is obtained as follows: ,in For the quality of drones, It is the acceleration due to gravity;

[0030] Step S22: Solve the ordinary differential equations of the ballistic descent under the influence of drag: The equations for the velocity changes in the horizontal and vertical directions are obtained as follows:

[0031]

[0032]

[0033] in, , , Changes are considered positive in the vertical upward direction; For the speed limit operator, , These are operators that limit the drone's motion direction after failure; all three are operator parameters.

[0034] Step S23: Calculate the four key time points of the drone crash, including the highest point time: Time of fall: Time of change in resistance direction:

[0035]

[0036] Total fall time:

[0037]

[0038] in, , where represents the upward displacement distance of the drone during its descent. , is the displacement direction constraint operator;

[0039] Step S24: Calculate the final drone crash probability density distribution:

[0040]

[0041] in, Given the heading angle, the expected average location of the impact point is:

[0042]

[0043] Using Taylor expansion, the analytical solution to the fall time equation is approximated as:

[0044]

[0045] The solution yields the two-dimensional probability density distribution function for the predicted impact point location:

[0046]

[0047] in, Finally, the two-dimensional distribution matrix of the fall locations within the required range is calculated. .

[0048] Furthermore, step S3 specifically includes:

[0049] Step S31: Calculate the kinetic energy of the drone upon impact: Obtain the two-dimensional distribution matrix of ground population density. ;

[0050] Step S32: Calculate the risk of personal injury under the influence of a drone accident: ,in, The fatality rate of personnel under the influence of impact kinetic energy is expressed as:

[0051]

[0052] in, J, J, , As the kinetic energy of the impact;

[0053] Step S33: Calculate the noise impact risk under the influence of drone accidents: ,in: The noise figure is calculated using the following formula:

[0054]

[0055] in, db, The noise attenuation index;

[0056] Step S34: Calculate the risk of property damage caused by a drone accident:

[0057]

[0058] in, , , ;

[0059] Step S35: Obtain the drone's in-flight failure probability matrix and accident impact risk matrix, and finally calculate the drone's instantaneous operational risk:

[0060]

[0061] in, The failure probability of the unmanned aerial vehicle system. The probability of collision between drones. The probability of a collision between a drone and an obstacle; the risk matrix of drone accident impact is as follows:

[0062]

[0063] in, The number of casualties is expressed in persons. The amount represents property loss, in yuan. Noise level, measured in decibels (dB).

[0064] The risks of instantaneous drone operation are:

[0065]

[0066] in, Risk coefficients under different accident impacts .

[0067] Compared with the prior art, the present invention has the following beneficial effects:

[0068] This invention fully considers the accident chain process of drone risk occurrence; it considers three in-flight failure scenarios: system failure probability, collision probability with drone, and collision probability with obstacle; and it solves the ballistic descent ordinary differential equation under the influence of wind resistance, which has universality. Attached Figure Description

[0069] Figure 1 This is a flowchart illustrating the overall implementation of an air-ground collaborative UAV operation risk assessment method according to the present invention. Detailed Implementation

[0070] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described below with reference to the accompanying drawings.

[0071] This invention provides a method for assessing the operational risks of unmanned aerial vehicles (UAVs) in an air-ground coordinated manner, such as... Figure 1 As shown, the specific steps are as follows:

[0072] Step 1: Calculate the failure probability of the UAV operating system and the probability of collision based on the system failure and collision probability density function distribution; specifically including:

[0073] Step 1.1: Based on system reliability theory, set the failure probability of the UAV system. It follows a Weibull distribution, and its probability density function is:

[0074]

[0075] in, This refers to the cumulative operating time of the drone. For proportional parameters, These are shape parameters, both determined by different stages of system operation.

[0076] Step 1.2: Obtain the initial nominal position of the UAV The trajectory error of the UAV is set to follow a Gaussian distribution, which is: ,in Let be the trajectory error values ​​along each direction of the aircraft's coordinate axes. Then the UAV's position distribution is:

[0077]

[0078] in, The nominal average value. This is the transformation matrix between the machine's coordinate axes and the global coordinate axes. denoted as the variance of the transformed probability distribution.

[0079] Step 1.3: Based on the above probability distribution definition, calculate the relative position of the UAV and the target obstacle: The relative position probability distribution of the two can be obtained as follows:

[0080]

[0081] in, The mean of relative positions. Let be the covariance of the probability distribution.

[0082] Step 1.4: Apply the track error distribution density function to calculate the collision probability between the UAV and the target obstacle.

[0083]

[0084] in, For runtime, For the integration region, Relative position The initial relative position mean, Let be the relative position covariance.

[0085] Step 2: Solve the probability density distribution function of the predicted crash point of the UAV under ballistic descent using a kinematic model; specifically including:

[0086] Step 2.1: Obtain the initial nominal position of the UAV With running speed drone quality Calculate the drag coefficient :

[0087]

[0088] in, air density, For windward area, Let be the air drag coefficient. The maximum descent speed of the drone is then obtained as: ,in For the quality of drones, This is the acceleration due to gravity.

[0089] Step 2.2: Solve the ordinary differential equations of the ballistic descent under the influence of drag: The equations for the velocity changes in the horizontal and vertical directions are obtained as follows:

[0090]

[0091]

[0092] in, , In the vertical direction, the upward direction is considered positive. For the speed limit operator, , These are the operators that limit the movement direction of the UAV after failure, and all three are operator parameters.

[0093] Step 2.3: Calculate the four key time points of the drone crash, including the highest point time: Time of fall: Time of change in resistance direction:

[0094]

[0095] Total fall time:

[0096]

[0097] in, , where represents the upward displacement distance of the drone during its descent. , is the displacement direction restriction operator.

[0098] Step 2.4: Calculate the final drone crash probability density distribution:

[0099]

[0100] in, Given the heading angle, the expected average location of the impact point is:

[0101]

[0102] Using Taylor expansion, the analytical solution to the fall time equation is approximated as:

[0103]

[0104] The solution yields the two-dimensional probability density distribution function for the predicted impact point location:

[0105]

[0106] in, Then the two-dimensional distribution matrix of the crash sites is: .

[0107] Step 3: Using an incident chain model, assess the instantaneous operational risk of the drone using multiple indicators; specifically including:

[0108] Step 3.1: Calculate the kinetic energy of the drone upon impact: Obtain the two-dimensional distribution matrix of ground population density: .

[0109] Step 3.2: Calculate the risk of personal injury under the influence of a drone accident: ,in: Fatality rate of people affected by impact kinetic energy.

[0110]

[0111] in, J, J, , This is the impact kinetic energy.

[0112] Step 3.3: Calculate the noise impact risk caused by drone accidents: ,in: The noise figure is calculated using the following formula:

[0113]

[0114] in, db, This is the noise attenuation index.

[0115] Step 3.4: Calculate the risk of property damage caused by a drone accident:

[0116]

[0117] in, , , .

[0118] Step 3.5: Obtain the drone's in-flight failure probability matrix and accident impact risk matrix, and finally calculate the drone's instantaneous operational risk:

[0119]

[0120] in, The failure probability of the unmanned aerial vehicle system. The probability of collision between drones. This represents the probability of a collision between the drone and an obstacle. The drone accident impact risk matrix is ​​as follows:

[0121]

[0122] in, The number of casualties is expressed in persons. The amount represents property loss, expressed in yuan. Noise level, measured in decibels.

[0123] The risks of instantaneous drone operation are:

[0124]

[0125] in, Risk coefficients under different accident impacts .

Claims

1. A method for assessing the operational risks of unmanned aerial vehicles (UAVs) in an air-ground coordinated manner, characterized in that, Includes the following steps: Step S1: Calculate the failure probability of the UAV operating system and the probability of collision failure based on the probability density function distribution of system failure and collision. Step S2: Solve the probability density distribution function of the expected crash point of the UAV under ballistic descent using a kinematic model; Step S3: Using an accident chain model, assess the instantaneous operational risk of the drone using multiple indicators, specifically including: Step S31: Calculate the kinetic energy of the drone upon impact: Obtain the two-dimensional distribution matrix of ground population density. ; Step S32: Calculate the risk of personal injury under the influence of a drone accident: ,in, The fatality rate of personnel under the influence of impact kinetic energy is expressed as: in, J, J, , As the kinetic energy of the impact; Step S33: Calculate the noise impact risk under the influence of drone accidents: ,in, The noise figure is calculated using the following formula: in, db, The noise attenuation index; Step S34: Calculate the risk of property damage caused by a drone accident: in, , , ; Step S35: Obtain the drone's in-flight failure probability matrix and accident impact risk matrix, and calculate the drone's instantaneous operational risk as follows: in, The failure probability of the unmanned aerial vehicle system. The probability of collision between drones. The probability of a collision between a drone and an obstacle; the risk matrix of drone accident impact is as follows: in, The number of casualties is expressed in persons. The amount represents property loss, in yuan. Noise level, measured in decibels (dB). The risks of instantaneous drone operation are: in, Risk coefficients under different accident impacts .

2. The method for assessing the operational risks of unmanned aerial vehicles (UAVs) in an air-ground coordinated manner according to claim 1, characterized in that, Step S1 specifically includes: Step S11: Based on system reliability theory, set the failure probability of the UAV system. It follows a Weibull distribution, and its failure probability function is: in, This refers to the cumulative operating time of the drone; For proportional parameters, These are shape parameters, both determined by different stages of system operation; Step S12: Obtain the initial nominal position of the UAV The trajectory error of the UAV is set to follow a Gaussian distribution, which is: ,in Let be the trajectory error values ​​along each direction of the aircraft's coordinate axes; then the position distribution of the UAV is: in, The nominal average value. This is the transformation matrix between the machine's coordinate axes and the global coordinate axes. The variance of the transformed probability distribution; Step S13: Based on the above probability distribution definition, calculate the relative position of the UAV and the target obstacle: in, The location of the drone. Location of the target obstacle. and These are the position transformation matrices for both. and These are the positional errors of the two, and These are the nominal positions of the two, respectively; Therefore, the probability distribution of their relative positions is obtained: in, The mean of relative positions. Let covariance be the probability distribution. Step S14: Calculate the collision probability between the UAV and the target obstacle by integrating the three-dimensional Gaussian distribution. in, This refers to the cumulative operating time of the drone. For the integration region, Relative position The initial relative position mean, Let be the relative position covariance.

3. The method for assessing the operational risks of unmanned aerial vehicles (UAVs) in an air-ground coordinated manner according to claim 1, characterized in that, Step S2 specifically includes: Step S21: Obtain the initial nominal position of the UAV With running speed drone quality Then calculate the drag coefficient. : in, air density, For windward area, The air drag coefficient is given; and the maximum descent speed of the UAV is obtained as follows: ,in For the quality of drones, It is the acceleration due to gravity; Step S22: Solve the ordinary differential equations of the ballistic descent under the influence of drag: The equations for the velocity changes in the horizontal and vertical directions are obtained as follows: in, , In the vertical direction, Changes are considered positive in the vertical upward direction; For the speed limit operator, , These are operators that limit the drone's motion direction after failure; all three are operator parameters. Step S23: Calculate the four key time points of the drone crash, including the highest point time: Time of fall: Time of change in resistance direction: Total fall time: in, , where represents the upward displacement distance of the drone during its descent. , is the displacement direction constraint operator; Step S24: Calculate the final drone crash probability density distribution: in, Given the heading angle, the expected average location of the impact point is: Using Taylor expansion, the analytical solution to the fall time equation is approximated as: The two-dimensional probability density distribution function of the predicted impact point is obtained by solving: in, Finally, the two-dimensional distribution matrix of the fall locations within the required range is calculated. .

Images